AI Paper Atlas

Research map / Probabilistic and scientific modeling / Causal discovery

Addressing divergent representations from causal interventions on neural networks

Satchel Grant, Simon Jerome Han, Alexa R. Tartaglini, Christopher Potts

ICLR 2026 · Probabilistic and scientific modeling › Causal discovery · approach: Causal representation identifiability · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add the Counterfactual Latent (CL) auxiliary loss to Boundless DASBoundless DAS on Llama-based LLM · Boundless DAS reproductionEMD and IIAImproved
train DAS with the CL loss instead of the behavioral loss onlyDAS on synthetic MLP · Synthetic held-out classesEMD: 0.032 -> 0.007Improved
train DAS with the CL loss instead of the behavioral loss onlyDAS on synthetic MLP · Synthetic held-out classes (test)IIA: 0.997 -> 0.9988No clear effect
train DAS with the CL loss instead of behavioral loss, evaluated on held-out partitionDAS alignment on synthetic dense/sparse partitions · Synthetic OOD transferOOD IIAImproved
higher intervened divergence (training EMD)DAS alignment · Synthetic OOD transferOOD IIA: coef. -.34, R^2=.73Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy